Passive Terahertz Imaging Research Articles

A novel framework for concealed weapons detection using passive terahertz imaging

ABSTRACT In this work, a novel framework for non-destructive, non-intrusive and completely automatic concealed weapon detection under human clothing for entry control and security check applications using passive terahertz (THz) images has been given. The technique models the problem of concealed weapon detection as that of binary segmentation, where the pixels corresponding to concealed weapons are foreground while the remaining are background. The core idea of the proposed framework is to first oversegment the THz image into superpixels and subsequently multiple handcrafted features are extracted from each superpixel. Finally, a machine learning-based classifier is used to perform binary classification thereby classifying each superpixel as foreground or background such that each pixel in a foreground superpixel is a foreground pixel and vice-versa for background. It is worth mentioning that this technique does not require a perfect segmentation to extract features from, rather an oversegmentation suffices. Furthermore, both global features like those based on image saliency and local features like those based on intensity are extracted considering the image properties. The detailed experiments and comparative analysis confirm that the proposed technique efficiently detects concealed weapons, exhibiting superior performance compared to state-of-the-art binary segmentation methods for the purpose.

Saliency and superpixel improved detection and segmentation of concealed objects for passive terahertz images

Application of passive terahertz imaging in concealed weapon detection has been looked at, such that the final result is the segmentation of the foreground concealed weapons from the rest of the background. For the same, a fully automatic and completely generic technique, without any learning, has been proposed. It was observed that a simple thresholding step, exploiting varied intensity bands of the tetrahertz images is not enough. Thus, an innovative method to isolate humans and thus improve the region of interest (ROI) has been proposed. Thereafter, saliency has been used to further improve ROI, as these images are quite noisy and the central focusing aspect of saliency could handle the noise around the concealed weapons. It was observed that this step could handle the noise around the concealed weapons but degraded the boundaries of the concealed weapons. To further improve boundary adherence, superpixels are used. Finally, results are evaluated both quantitatively and qualitatively and outperformed the traditional approach.

Multipolarization Image Evaluation for Passive Terahertz Imaging Detection

Multipolarization Image Evaluation for Passive Terahertz Imaging Detection

340 and 250 GHz Schottky solid‐state heterodyne receiver arrays for passive imaging systems

AbstractReal‐time terahertz passive imaging systems are highly demanded in security checking application. In this paper, two four‐pixel receiver arrays for 340 and 250 GHz passive imaging systems are presented. As the core device of the front‐end of the receiving array, the low noise terahertz mixer is realized based on the enhanced Schottky diode model and low‐loss circuit structure. The minimum double side band (DSB) noise temperature of the separated 340 and 250 GHz mixers is 900 and 580 K, respectively, within the operational band‐width. Without considering fluctuation in the power gain, the calculated temperature sensitivity of 340 GHz receiver is 0.8 K with 40 GHz RF band. Finally, based on all‐solid‐state millimeter‐wave drive front end and reflector optics, a passive terahertz imaging system is completed. Less than 1 cm space resolution is achieved for 1–3 m object distances, and the distinct imaging to hidden threats is produced and made contrast among the two receiver arrays. The perfect imaging performances of the system verify the low noise characteristic of terahertz mixer. The Schottky solid‐state receiver array provides a cheap and feasible option for personal security screening.

YOLO-MSFG: Toward Real-Time Detection of Concealed Objects in Passive Terahertz Images

Passive terahertz (THz) systems can perform the perspective imaging of objects concealed underneath clothing on the human body without a radiometric emitter. In this study, a linear-array passive THz imaging system was designed to realize highly efficient push-broom imaging. However, the acquired THz images are contaminated by severe striping and random noise, which greatly hampers object detection. Therefore, this paper aims to report a dedicated deep learning network for the automatic, accurate, and real-time detection of concealed objects in passive THz images. The main work of this research is threefold. First, a bilateral filter is integrated into a convolutional neural network (CNN) to generate a space-range tunable learning architecture. Second, a space-range grid combined with a geometric transformation matrix is designed to provide a core unit for multi-scale filtering and geometric (MSFG) augmentation. Finally, the one-stage YOLOv5 detector is improved to YOLOv5I by refining the granularity of its prediction units. The proposed approach was tested on onsite passive THz images acquired from various scenes, and the results demonstrate that the proposed MSFG augmentation method can significantly improve the detection accuracy of all the candidate CNNs on the passive THz images. Among them, the YOLOv5I-MSFG was found to exhibit the best performance via a comprehensive assessment of its real-time ability and detection accuracy; the time consumption per frame was found to be <inline-formula> <tex-math notation="LaTeX">$\sim 53$ </tex-math></inline-formula> ms on the configured platform, and the accuracies of one-class and two-class detection were found to reach 93.49&#x0025; and 90.76&#x0025;, respectively.

AC‐SDBSCAN: Toward concealed object detection of passive terahertz images

The passive terahertz (THz) detection on objects concealed under clothing is quite challenging due to the heavy noise interference. In this paper, an active-contour-based self-adaptive DBSCAN (AC-SDBSCAN) detection algorithm is proposed. The core of AC-SDBSCAN algorithm lies in that the object contours are first extracted by AC method in a noise scenario and then the statistical features of the contours are used to motivate a SDBSCAN to complete clustering without initialization. Benefiting from the strong robustness of AC and DBSCAN to the noise, the concealed objects in the noisy THz images can be detected accurately. Extensive simulations are verified on four passive THz image datasets. The results indicate that the AC method in our solution can achieve over 87% accuracy for contour extraction of passive THz images, while the classical methods achieve less than 77%; in addition, the SDBSCAN can achieve over 90% clustering accuracy without manual initialization which is significantly superior to the conventional DBSCAN. Eventually, the proposed method completes the object detection of passive THz images with a maximum recall of 90.38% and a maximum precision of 94%.

Wideband Modeling of CMOS Schottky Barrier Diode Detectors for THz Radiometry

A complete system modeling and characterization of a wideband differential terahertz (THz) direct detector, integrated in a commercial CMOS technology, is presented. The detector consists of a recently developed double leaky-slot lens antenna that operates from 200 to 600 GHz in combination with a differential Schottky barrier diode (SBD) direct detection circuit. The proposed methodology, starting from low-frequency measurements on a standalone SBD, is able to adequately model the spectral radiometric performance. The system noise-equivalent power (NEP) is characterized from 325 to 500 GHz in excellent agreement with the proposed system model. The measured NEP, 20 pW/√Hz minimum and 90 pW/√Hz frequency averaged, is compromised with respect to the average NEP of 2.7 pW/√Hz that was initially predicted by simulations using the process design kit (PDK) model, since the available SBDs are operating beyond their cutoff frequency. The diodes and models provided by the PDK proved to be inaccurate in predicting circuit behavior at these high frequencies. By using the proposed analysis and modeling approaches, an accurate wideband antenna-detector codesign could be applied for future passive THz imaging applications based on CMOS technologies.

Roadmap of Terahertz Imaging 2021.

In this roadmap article, we have focused on the most recent advances in terahertz (THz) imaging with particular attention paid to the optimization and miniaturization of the THz imaging systems. Such systems entail enhanced functionality, reduced power consumption, and increased convenience, thus being geared toward the implementation of THz imaging systems in real operational conditions. The article will touch upon the advanced solid-state-based THz imaging systems, including room temperature THz sensors and arrays, as well as their on-chip integration with diffractive THz optical components. We will cover the current-state of compact room temperature THz emission sources, both optolectronic and electrically driven; particular emphasis is attributed to the beam-forming role in THz imaging, THz holography and spatial filtering, THz nano-imaging, and computational imaging. A number of advanced THz techniques, such as light-field THz imaging, homodyne spectroscopy, and phase sensitive spectrometry, THz modulated continuous wave imaging, room temperature THz frequency combs, and passive THz imaging, as well as the use of artificial intelligence in THz data processing and optics development, will be reviewed. This roadmap presents a structured snapshot of current advances in THz imaging as of 2021 and provides an opinion on contemporary scientific and technological challenges in this field, as well as extrapolations of possible further evolution in THz imaging.

Reflection Removal Using Dual-Polarization and Saliency in Millimeter-Wave and Terahertz Imaging

Passive millimeter-wave and terahertz (MMW/THz) imaging is a significant technique used in personal security inspection and scene surveillance. However, a human body generally generates an obvious reflection on the floor, which may have negative influences on target detection and observer vision. In this article, we present a physically based method using dual-polarization feature and saliency information to solve the problem of reflection removal. The physical principle of a special dual-polarization phenomenon has been revealed by theoretical calculation and simulation analysis. By combining the second Stokes parameter (TQ) and modified linear polarization ratio (mLPR), a new feature parameter has been created to extract reflections of human bodies. By using a saliency analysis, the false positive alarms are eliminated completely. The experimental results demonstrate the robustness and effectiveness of the proposed method using various measured MMW/THz images.

Passive imaging at 250 GHz for detection of face presentation attacks.

Face presentation attacks are becoming more efficient since new 3D facial masks are used. Passive terahertz imaging offers specific physical properties that may improve presentation attack detection capabilities. The non-zero transmission capability through a variety of non-metallic materials may provide necessary information for presentation attack detection. The aim of this paper is to present outcomes of a study on face presentation attack detection using passive imaging at 250 GHz. An analysis of presentation attacks for facial recognition systems using custom displayed and printed photographs, 3D-printed and full-face flexible 3D-latex masks, is provided together with spectral characterization of various presentation attack instruments. A set of experiments with various instruments and various sets of clothing is described and discussed. Finally, two presentation attack detection methods are proposed. The first method is based on a threshold corresponding to a difference between mean intensities of selected regions of interests while the second method uses eight different deep learning classifiers to detect presentation attacks. Results of two validation schemes are presented.

A passive video-rate terahertz human body imager with real-time calibration for security applications

Real-time video-rate passive terahertz imaging systems are highly demanded for practical applications, especially in security checking. Here, we demonstrate a passive video-rate terahertz human body imaging system, which was mainly consisted of a scanning module, a quasi-optical lens, a calibration module and a one-dimensional terahertz detector array. The terahertz waves radiated from human bodies in front of the imager can transmit through a terahertz window into the imager, and were reflected by the scanning module, and then focused on the detector array by the quasi-optical lens. A calibration module was also designed to calibrate the terahertz detectors in real-time without disturbing the imaging process. In combination of the scanning module with the detector array, the imager can obtain a full image of a human body standing at a distance of 1.5 m in front of the imager with a resolution of 1.5 cm and a frame rate of 10 fps. The imaging system can discover suspected dangerous items carried on the human body such as metals, ceramics, powders and liquids. Furthermore, an intelligent terahertz imaging algorithm employing convolutional neural network was also successfully realized based on the terahertz images produced by this system to improve the image quality and mark the detected items automatically. We believe our real-time video-rate terahertz imaging techniques and systems not only have great values for further inspiring developing terahertz imaging systems but also can accelerate the terahertz technology towards more practical applications.

Passive terahertz imaging detectors based on antenna-coupled high-electron-mobility transistors.

Aiming at the requirement of passive terahertz imaging, we report a high-sensitivity terahertz detector based on an antenna-coupled AlGaN/GaN high-electron-mobility transistor (HEMT) at 77 K without using low-noise terahertz amplifier. The measured optical noise-equivalent power and the noise-equivalent temperature difference of the detector were about 0.3p W/H z and 370 mK in a 200 ms integration time over a bandwidth of 0.7 - 0.9 THz, respectively. By using this detector, we demonstrated passive terahertz imaging of room-temperature objects with signal-to-noise ratio up to 13 dB. Further improvement in the sensitivity may allow passive terahertz imaging using AlGaN/GaN-HEMT at room temperature.

基于被动式人体安检太赫兹图像的小目标物体检测方法

基于被动式人体安检太赫兹图像的小目标物体检测方法

单探测器快速扫描被动太赫兹人体成像系统

单探测器快速扫描被动太赫兹人体成像系统

Multiplexed readout of kinetic inductance bolometer arrays

Kinetic inductance bolometer technology is a candidate for passive submillimeter wave and terahertz imaging systems. Its benefits include scalability into large 2D arrays and operation with intermediate cryogenics in the temperature range of 5-10 K. We have previously demonstrated the scalability in terms of device fabrication, optics integration, and cryogenics. In this article, we address the last missing ingredient, the readout. The concept, serial addressed frequency excitation, is an alternative to full frequency-division multiplexing at microwave frequencies conventionally used to read out kinetic inductance detectors. We introduce the concept and analyze the criteria of the multiplexed readout avoiding the degradation of the signal-to-noise ratio in the presence of a thermal anti-alias filter inherent to thermal detectors. We present a practical scalable realization of a readout system integrated into a prototype imager with 8712 detectors. This is used for demonstrating the noise properties of the readout. Furthermore, we present practical detection experiments with a stand-off laboratory-scale imager.

Using Markov constraint and constrained least square filter to develop a novel method of passive terahertz image restoration

In recent years, passive terahertz imaging has gained significant attention in both research and practice. One big challenge with passive terahertz imaging is its low-quality images with high level of noise. State-of-the art image restoration methods have been developed for image denoising, such as methods based on Markov constraint and regular filter methods. Building upon these two methods, this paper develops a novel method for passive terahertz image restoration which preserves well both high frequency and low frequency information of the images. Performance of our method is evaluated using two common image criteria of the image sharpness, i.e. edge intensity and definition. Experimental results showed our method outperform state-of-the art methods for passive terahertz image restoration.

Terahertz / Visible Dual-band Image Fusion Based on Hybrid Principal Component Analysis

This study puts forward a kind of passive terahertz image and visible image fusion method. Our method takes two steps of image fusion. First step involves a pre-processing of the passive terahertz image and image fusion using principal component analysis. Registered terahertz image is segmented and pseudo-color encoded. Taking the characteristics of terahertz image and visible image into account, principal component analysis is used to fuse the dualband images. HIS method is used for the second step of image fusion. The experiment results show that this algorithm can effectively fuse passive terahertz and visible image and is helpful to detect and locate dangerous objects concealed under the clothes of the target subjects to strengthen the practicability of terahertz imaging system for security screening.

Study of Automatic Detection of Concealed Targets in Passive Terahertz Images for Intelligent Security Screening

The automatic extraction of the targets in which we are interested from a given image is the fundamental of the automatic detection and identification for security screening systems based on imaging technologies. Suffering from the comparatively low signal-to-noise ratio (SNR), the automatic detection of targets in a passive terahertz (THz) imaging system facing great challenges, but in urgent necessary. In this paper, a comprehensive method for automatic detection of concealed targets in passive THz image by making the best use of the “block statistics uniformity” properties of the passive images is first studied. A theoretical model for the “featured regions” decomposition based on the minimization of a “fit energy” functional with respect to a “surface function” is established, to overcome the drawbacks of conventional methods with gradient-based edge operators for their unsuccessful application in low SNR passive images with blurred boundaries. Based on earlier theoretical basis and taking advantages of the distinguished contrasts of the convergent “surface function” in different “featured regions,” an automatic detection algorithm with three steps was further developed to automatically extract the number, the locations and the shapes of all the concealed targets, with the shape of each target derived as the contour point series arranged in clockwise direction. With plenty of experimental results in 0.2 THz band, it is found that, the proposed method has high detection accuracy about 95% with quite good realtime performance, even for the single channel proof-of-state system with low SNR. The theorem, algorithm, and results, in this paper, may have important applications in unmanned and intelligent security screening systems without any artificial interventions.

Design and Characterisation of Titanium Nitride Subarrays of Kinetic Inductance Detectors for Passive Terahertz Imaging

We report on the investigation of titanium nitride (TiN) thin films deposited via atomic layer deposition (ALD) for microwave kinetic inductance detectors (MKID). Using our in-house ALD process, we have grown a sequence of TiN thin films (thickness 15, 30, 60 nm). The films have been characterised in terms of superconducting transition temperature T_mathrm{c}, sheet resistance R_mathrm{s} and microstructure. We have fabricated test resonator structures and characterised them at a temperature of 300 mK. At 350 GHz, we report an optical noise equivalent power hbox {NEP}_mathrm{opt} approx 2.3times 10^{-15}~hbox {W}/sqrt{hbox {Hz}}, which is promising for passive terahertz imaging applications.

Terahertz Imaging Progress at Capital Normal University

The progress on terahertz imaging at Capital Normal University in Beijing is presented. Our works on Terahertz Imaging include the active and passive imaging. For the active terahertz imaging, the pulse and continue wave terahertz imaging are studied, respectively. The active terahertz pulse imaging is based on the terahertz time-domain spectroscopy (THz-TDS). A practical focus-plane imaging system is built up based on the THz-TDS with the probe-beam-expanded electro-optical sampling and an infrared CCD imaging. The active terahertz continuous wave imaging is based on a CO₂-laser-pumped terahertz coherent source and an uncooled bolometer array of terahertz camera. The active terahertz polarization imaging and wave front imaging is developed. For the passive terahertz imaging, the low frequency of radiometers are used to detect the point-to-point beam-scanned terahertz signal by the optical focusing and scanning system so that the passive THz imaging is realized for the security inspection of human body. The related components and image processing methods are studied and used for the improvement of imaging speed and resolution. The research on THz imaging technology at Capital Normal University is opening up a potential application in the field of non-destructive testing and security inspection for the terahertz technology.